"Uranium" required 13.2 " 10.0 " 6.0 " 2.0 "

These show that the quant.i.ty ordered (5 c.c.) must be adhered to.

~Effect of Foreign Salts.~--In these experiments, 10 grams of the salt (the effect of which it was desired to determine) were added to a solution in other respects resembling those previously used:--

Salt added {Ammonic Ammonic Ammonic Magnesium {sulphate nitrate chloride sulphate "Uranium" required 15.5 c.c. 15.5 c.c. 15.3 c.c. 15.3 c.c.

Without any addition, 15.0 c.c. were required; and in another experiment, in which 30 grams of ammonic salts were present, 15.6 c.c.

of uranium solution were required. Such variations in the amount of ammonic salts as occur in ordinary working are unimportant.

Phosphates, of course, interfere. In fact, the uranium acetate solution can be standardised by t.i.trating with a known weight of phosphate, and calculating its equivalent of a.r.s.enic. Thus, in an experiment with 0.6 gram of hydric sodic phosphate (Na_{2}HPO_{4}.12H_{2}O), equivalent to 0.05195 gram of phosphorus, or 0.1256 gram of a.r.s.enic, 23.25 c.c. of a solution of uranium acetate were required. The same solution standardised with white a.r.s.enic gave a standard of which 100 c.c. = 0.5333 gram a.r.s.enic. On this standard the 0.6 gram of sodic phosphate should have required 23.5 c.c.

Experiments in which 0.1 gram of bis.m.u.th and 0.1 gram of antimony were present with 0.1 gram of a.r.s.enic, showed no interference on the t.i.tration. Ferric or aluminic salts would remove their equivalent of a.r.s.enic, and, consequently, must be removed before t.i.trating.

~Effect of Varying a.r.s.enic.~--Varying amounts of metallic a.r.s.enic were weighed up and dissolved in nitric acid, &c., and t.i.trated:--

a.r.s.enic taken 0.010 gram 0.050 gram 0.100 gram 0.200 gram a.r.s.enic found 0.010 " 0.050 " 0.100 " 0.197 "

These experiments show that the method yields good results within these limits.

~Determination of a.r.s.enic in Mispickel.~--Weigh up 1 gram of the dried and powdered ore, and evaporate to near dryness with 20 c.c. of dilute nitric acid. Make up to 100 c.c. with water, and pa.s.s sulphuretted hydrogen to reduce the ferric iron to the ferrous state, then add 20 c.c. of dilute ammonia, and again pa.s.s sulphuretted hydrogen. Warm, filter, and evaporate the filtrate to drive off the excess of ammonia; then add 10 c.c. of nitric acid, and boil down till the sulphide of a.r.s.enic at first precipitated is dissolved; neutralise; add 5 c.c. of sodium acetate and acetic acid solution; transfer to a pint flask, boil, and t.i.trate.

For example, an impure sample of ore required, in duplicate a.s.say of half a gram each, when treated in the above-mentioned way, 39.6 and 39.5 c.c. of the uranium acetate solution (100 c.c. = 0.537 gram of a.r.s.enic), equivalent to 0.2114 gram of a.r.s.enic, or 42.3 per cent.

An alternative method is as follows. Powder the ore very finely and weigh up .5 gram. Place in a 2-3/4 inch berlin dish and add strong nitric acid, one drop at a time until the action ceases; with care there need be no very violent reaction. Dry over a water bath. Cover with 2 grams of nitre and over this spread 5 grams of a mixture of equal parts of nitre and carbonate of soda. Fuse in a m.u.f.fle or over a large gentle blow-pipe flame for 4 or 5 minutes. This will spoil the dish. Allow to cool and boil out in a larger dish with 100 c.c. of water. Filter and wash into an 8 oz. flask. Acidify the liquor with nitric and boil down to about 100 c.c. The acid should not be in too large excess, but an excess is needed to destroy nitrites. Neutralise with soda or ammonia.

Add 5 c.c. of the mixture of sodium acetate and acetic acid. t.i.trate with uranium acetate.

~Determination of a.r.s.enic (As) in Crude a.r.s.enic.~--The method given under the iodine t.i.tration simply determines that portion of the a.r.s.enic which is present in the substance as a.r.s.enious oxide or white a.r.s.enic.

The following method will give the total a.r.s.enic in the sample. It would be incorrect to report this as so much per cent. of a.r.s.enious oxide, although it may be reported as so much per cent. of a.r.s.enic equivalent to so much per cent. of white a.r.s.enic, thus:--

a.r.s.enic 30.0 per cent.

Equivalent to white a.r.s.enic 39.6 "

The equivalent of white a.r.s.enic is calculated by multiplying the percentage of a.r.s.enic by 1.32. The method of determining the percentage of a.r.s.enic is as follows:---Boil 1 gram of the sample with 10 c.c. of nitric acid. When the bulk of the solution has been reduced to one-half, and red fumes are no longer evolved, dilute with a little water, and filter into a flask. Neutralise the filtrate, add 5 c.c. of sodic acetate solution, boil and filter. The precipitate (ferric a.r.s.enate) is transferred to a small beaker, treated with 5 c.c. of dilute ammonia, and sulphuretted hydrogen pa.s.sed through it. The iron sulphide is filtered off, and the filtrate evaporated with an excess of nitric acid.

When the solution is clear, it is neutralised, and 1 or 2 c.c. of sodic acetate solution having been added, is then mixed with the first filtrate. The solution is boiled and t.i.trated.

A sample treated in this way required 49.2 c.c. of the uranium acetate solution (100 c.c. = 0.537 gram of a.r.s.enic), equivalent to 26.4 per cent.

~Determination of a.r.s.enic in Brimstone.~--Take 10 grams of the substance, and powder in a mortar; rub up with 10 c.c. of dilute ammonia and a little water; rinse into a pint flask; pa.s.s a current of sulphuretted hydrogen; and warm on a hot plate for a few minutes.

Filter, acidulate the filtrate with sulphuric acid; filter off the precipitate; attack it with 10 c.c. of nitric acid; and proceed as in the other determinations.

PRACTICAL EXERCISES.

1. Mispickel contains 45.0 per cent. of a.r.s.enic, to how much white a.r.s.enic will this be equivalent?

2. How would you make a standard solution of iodine so that 100 c.c.

shall be equivalent to 1 gram of white a.r.s.enic?

3. What weight of a.r.s.enic is contained in 1 gram of pyra.r.s.enate of magnesia, and what weight of ammonic-magnesic a.r.s.enate would it be equivalent to?

4. The residue, after heating 10 grams of crude a.r.s.enic, weighed 0.62 gram. What information does this give as to the composition of the substance? If another 10 grams of the substance, heated on a water-bath, lost 0.43 gram, what conclusions would you draw, and how would you report your results?

5. If a sample of copper contained 0.5 per cent. of a.r.s.enic, and 1 gram of it were taken for an a.s.say, how much standard uranium acetate solution would be required in the t.i.tration?

PHOSPHORUS AND PHOSPHATES.

Phosphorus rarely occurs among minerals except in its highest oxidized state, phosphoric oxide (P_{2}O_{5}), in which it occurs abundantly as "rock phosphate," a variety of apat.i.te which is mainly phosphate of lime. Phosphates of most of the metallic oxides are found. Phosphoric oxide in small quant.i.ties is widely diffused, and is a const.i.tuent of most rocks. Its presence in varying amounts in iron ores is a matter of importance, since it affects the quality of the iron obtainable from them.

Phosphorus occurs in alloys in the unoxidized state. It is directly combined with the metal, forming a phosphide. In this manner it occurs in meteoric iron. The alloy phosphor-bronze is made up of copper, tin, zinc, and phosphorus.

Phosphates are mined in large quant.i.ties for the use of manure manufacturers, and for making phosphorus.

Phosphorus and a.r.s.enic closely resemble each other in their chemical properties, more especially those which the a.s.sayer makes use of for their determination. Phosphorus forms several series of salts; but the phosphates are the only ones which need be considered. Pyrophosphate of magnesia, which is the form in which phosphoric oxide is generally weighed, differs from the ordinary phosphate in the proportion of base to acid. Metaphosphates differ in the same way. If these are present, it must be remembered they act differently with some reagents from the ordinary phosphates, which are called orthophosphates. They are, however, all convertible into orthophosphates by some means which will remove their base, such as fusion with alkaline carbonates, boiling with strong acids, &c.[108]

Phosphides are converted into phosphates by the action of nitric acid or other oxidizing agents. Dilute acids, when they act on the substance, evolve phosphuretted hydrogen (PH_{3}). The student should be on his guard against losing phosphorus in this manner.

There is no dry a.s.say for phosphorus. All a.s.says for it are made either gravimetrically or volumetrically.

The separation of phosphoric oxide is made as follows:--The ore or metal is dissolved in acid and evaporated, to render the silica insoluble. It is taken up with hydrochloric acid, diluted with water, and treated with sulphuretted hydrogen. The filtrate is boiled, to get rid of the excess of gas, and treated with nitric acid, to peroxidize the iron present. If the iron is not present in more than sufficient quant.i.ty to form ferric phosphate with all the phosphorus present, some ferric chloride is added. The iron is then separated as basic acetate. The precipitate will contain the phosphorus, together with any a.r.s.enic acid not reduced by the sulphuretted hydrogen. The precipitate should have a decided brown colour. The precipitate is washed, transferred to a flask, and treated first with ammonia, and then with a current of sulphuretted hydrogen.

The filtrate from this (acidulated with hydrochloric acid, and, if necessary, filtered) contains the phosphorus as phosphoric acid. This method is not applicable in the presence of alumina, chromium, t.i.tanium, or tin, if the solution is effected with nitric acid. The precipitate obtained by the action of nitric acid on tin retains any phosphoric or a.r.s.enic oxide that may be present.

A method of separation more generally applicable and more convenient to work is based on the precipitation of a yellow phospho-molybdate of ammonia,[109] by the action of an excess of ammonic molybdate upon a solution of a phosphate in nitric acid. Dissolve the substance by treatment with acid, and evaporate to dryness. Take up with 10 c.c. of nitric acid, and add 20 grams of ammonic nitrate, together with a little water. Next put in the solution of ammonium molybdate solution in the proportion of about 50 c.c. for each 0.1 gram of phosphoric oxide judged to be present. Warm to about 80 C., and allow to stand for an hour.

Filter, and wash with a 10 per cent. solution of ammonic nitrate. It is not necessary that the whole of the precipitate be placed on the filter; but the beaker must be completely cleaned. Dissolve the precipitate off the filter with dilute ammonia, and run the solution into the original beaker. Run in from a burette, slowly and with stirring, "magnesia mixture," using about 15 c.c. for each 0.1 gram of phosphoric oxide.

Allow to stand for one hour. The white crystalline precipitate contains the phosphorus as ammonium-magnesium phosphate.

Phosphate of lead is decomposed by sulphuric acid; the lead is converted into the insoluble lead sulphate, and the phosphoric acid is dissolved.

Phosphate of copper and phosphate of iron may be treated with sulphuretted hydrogen; the former in an acid, and the latter in an alkaline, solution. Phosphate of alumina is generally weighed without separation of the alumina, since this requires a fusion. In all cases the aim is to get the phosphoric oxide either free, or combined with some metal whose phosphate is soluble in ammonia.

Joulie"s method of separation is as follows:--One to ten grams of the sample are treated with hydrochloric acid, and evaporated to dryness with the addition (if any pyrites is present) of a little nitric acid.

The residue is taken up with hydrochloric acid, cooled, transferred to a graduated flask, and diluted to the mark. It is then shaken up, filtered through a dry filter, and a measured portion (containing about 0.05 gram of phosphoric acid) transferred to a small beaker. Ten c.c. of a citric-acid solution of magnesia[110] is added, and then an excess of ammonia. If an immediate precipitate is formed, a fresh portion must be measured out and treated with 20 c.c. of the citrate of magnesia solution and with ammonia as before. The beaker is put aside for from two to twelve hours. The precipitate is then filtered off and washed with weak ammonia; it contains the phosphorus as ammonium-magnesium phosphate.

GRAVIMETRIC DETERMINATION.

If the phosphate is not already in the form of ammonic-magnesic phosphate, it is converted into this by the addition to its solution of an excess of ammonia and "magnesia mixture." In order to get the precipitate pure, the "magnesia mixture" is run in gradually (by drops) from a burette, with constant stirring. A white crystalline precipitate at once falls, if much phosphorus is present; but, if there is only a small quant.i.ty, it may be an hour or two before it shows itself. The solution is best allowed to rest for twelve or fifteen hours (overnight) before filtering. The presence of tartaric acid should be avoided; and the appearance of the precipitate should be crystalline. The solution is decanted through a filter, and the precipitate washed with dilute ammonia, using as little as may be necessary. The precipitate is dried, transferred to a weighed Berlin or platinum crucible; the filter-paper is carefully burnt, and its ash added to the precipitate, which is then ignited, at first gently over a Bunsen burner, and then more strongly over the blowpipe or in the m.u.f.fle. The residue is a white ma.s.s of magnesium pyrophosphate containing 27.92 per cent. of phosphorus, or 63.96 per cent. of phosphoric oxide.

VOLUMETRIC METHOD.

Instead of separating and weighing this compound, the phosphoric oxide in it can be determined by t.i.tration. In many cases the ore may be dissolved and immediately t.i.trated without previous separation. It is better, however, to carry the separation so far as to get phosphoric acid, an alkaline phosphate, or the magnesia precipitate. It may then be prepared for t.i.tration in the following way:--The precipitate in the last case (without much washing) is dissolved in a little hydrochloric acid, and the solution in any case rendered fairly acid. Dilute ammonia is added till it is just alkaline, and then 5 c.c. of the sodic acetate and acetic acid mixture (as described under the a.r.s.enic a.s.say). This should yield a clear distinctly-acid solution. It is diluted to 100 or 150 c.c., heated to boiling, and t.i.trated with the uranium acetate solution, using that of pota.s.sic ferrocyanide as indicator.

The _standard solution_ required is made by dissolving 35 grams of uranium acetate in water with the aid of 25 c.c. of acetic acid, and diluting to 1 litre.

An _equivalent solution of phosphoric oxide_ is made by dissolving 25.21 grams of crystallised hydric disodic phosphate (HNa_{2}PO_{4}.12H_{2}O) in water, and making up to 1 litre. 100 c.c. will contain 0.5 gram of phosphoric oxide (P_{2}O_{5}), or 0.2183 gram of phosphorus. In making this solution, transparent crystals only must be used. The uranium acetate solution is only approximately equivalent to this, so that its exact standard must be determined.

_Sodic Acetate and Acetic Acid Solution._--It is the same as that described under _a.r.s.enic_.[111] Use 5 c.c. for each a.s.say.